Classroom Technology Research for Kids: What the Studies Actually Show

The pitch arrives in every school newsletter, every district technology plan, and every Parent Teacher Organization meeting: technology is transforming education, and schools that don’t adopt the latest tools are leaving students behind. It sounds reasonable. It’s backed by billion-dollar marketing budgets. And when you look at the actual research on classroom technology for kids — the randomized controlled trials, the OECD analyses, the long-term outcome studies — the picture that emerges is considerably more complicated than the sales deck suggested.

Key Takeaways

• The largest cross-national study of classroom technology (OECD, 2015) found that higher computer use in school was associated with worse reading and math outcomes in most countries studied.
• Tablets and educational apps show modest benefits in narrow contexts — mainly for drill-and-practice tasks, early phonics, and specific math skills — but do not outperform well-designed traditional instruction in most comparisons.
• Smartboards and interactive whiteboards show no consistent academic advantage over standard whiteboards in rigorous studies; teacher skill matters far more than the technology itself.
• Implementation quality, teacher training, and how devices are integrated into instruction determine outcomes more than the technology type.
• The strongest evidence for any edtech tool currently is in adaptive learning software for math in specific grade bands — and even there, effects are modest.

The Problem: $13 Billion in Search of Evidence

American schools spent an estimated $13 billion on education technology in 2022 alone. iPad programs, one-to-one laptop initiatives, interactive whiteboards, smartboard systems, learning management platforms, reading apps, math apps — the investment has been enormous and accelerating, particularly since the pandemic created both urgency and federal funding through the ESSER relief program.

The problem is that the evidence base for most of these investments ranges from thin to nonexistent. EdTech companies routinely cite their own internal studies — typically observational comparisons without control groups, run over short time periods, conducted on motivated early adopters. These studies almost never appear in peer-reviewed journals. When independent researchers run the same comparisons with proper controls, the results consistently fail to replicate the company-reported benefits.

This isn’t a marginal finding. It’s a structural feature of how edtech adoption decisions get made. School boards and district technology committees are evaluating purchase decisions worth millions of dollars using evidence standards that would be rejected by any peer-reviewed journal. A company that can show a 15% improvement in reading fluency scores in a six-week pilot study — with no control group, no pre-registration, and a sample of students whose teachers were self-selected enthusiasts — can often get a district contract. The same study, submitted to the Journal of Educational Psychology, would be rejected without revision rights.

Parents often assume that if the district adopted a technology platform, someone evaluated the evidence. They usually didn’t — or evaluated it using the company’s materials. Understanding what the independent research actually shows gives parents a better basis for the conversations worth having.

What the Research Actually Says

The OECD 2015 Report: Students, Computers and Learning

This is the most important and most frequently misrepresented study in the edtech debate. The Organisation for Economic Co-operation and Development analyzed data from the 2012 PISA assessments — involving more than 900,000 students across 70 countries — to examine the relationship between computer use in school and academic performance.

The findings were striking and largely negative. Countries where students reported greater use of computers in school showed no improvement in reading or math scores, and in most cases showed lower scores than countries with lower computer use. Students who used computers moderately (several times per week) performed better than those who used them daily — suggesting a diminishing-returns or even negative relationship above a threshold. The report’s authors were careful to note that correlation isn’t causation, and that some of the variation reflects which subjects computers were used for and how. But the finding directly contradicted the assumption that more technology = better learning. The OECD explicitly stated that “investing heavily in school computers and connectivity is not linked to improved student achievement” based on this data.

The 2015 report was widely reported and then widely ignored by the edtech industry, which continued marketing products using weaker internal studies. Most district technology plans written after 2015 don’t cite it.

Cheung and Slavin’s Meta-Analysis on Educational Technology (2013)

Alan Cheung and Robert Slavin at Johns Hopkins conducted a rigorous systematic review of research on technology-based instruction for math and reading in K-12, published in the Educational Research Review. They identified 84 studies meeting basic quality standards (though most were not randomized controlled trials). Their findings: the average effect size across all technology interventions was 0.16 for math and 0.16 for reading — positive but modest, roughly equivalent to a few additional months of learning per year, and smaller than the effects seen from well-implemented tutoring, feedback, or cooperative learning programs.

Critically, they found significant variation. The highest effects were for supplemental programs — technology used in addition to, not instead of, regular instruction. Programs that replaced teacher-led instruction with technology showed smaller effects. This finding has been replicated in multiple subsequent reviews: technology works best as a supplement that provides practice opportunities, not as a replacement for direct instruction.

Randomized Controlled Trials on Tablets and iPads

The experimental literature on tablets is small but informative. A 2019 RCT published in AERA Open by Zheng and colleagues examined whether providing students with iPads improved learning outcomes in elementary reading and math. The study found no significant differences in standardized reading or math achievement between students who received iPads and those who didn’t, after one full academic year. Engagement and student attitudes toward school were slightly higher in the iPad group — suggesting tablets may affect motivation without affecting measured learning.

A series of UK studies on interactive whiteboards (Moss et al., 2007; and subsequent replication work) found that smartboards produced no consistent improvement in reading, math, or science achievement compared to standard whiteboards. Teacher enthusiasm for the technology was high and teacher confidence increased — but this teacher confidence effect didn’t translate into student outcome differences. The researchers concluded that the quality of the lesson determined the outcome, not the presence of the technology.

Adaptive Learning Software: The Strongest Case for EdTech

The most consistent positive evidence in the classroom technology literature is for adaptive learning software — programs that adjust difficulty and content in real time based on student performance. DreamBox Learning (math), Khan Academy, and similar platforms have been evaluated in studies with better design than most edtech research.

A 2016 study published in the Journal of Research on Educational Effectiveness examined DreamBox use in elementary math across multiple schools and found an average effect size of 0.17–0.22 in math fluency for students who used the program as a supplement, at recommended dosages (20–30 minutes, 2–3 times per week). These are modest but real effects. Importantly, they were not found for students who used the program less frequently than recommended, or for students in schools where the program replaced rather than supplemented instruction.

The honest interpretation of the adaptive learning literature: these tools can produce meaningful learning gains in specific skill domains (primarily math fact fluency and early phonics) under specific conditions (supplemental, used at recommended dosage, with teacher oversight). They’re not transformative. They’re a reasonable tool for the right job.

What Doesn’t Work: Apps as Learning Surrogates

The consumer education app market — apps sold to parents as learning tools for toddlers through middle school — has essentially no credible evidence of academic benefit. A 2019 study by Hirsh-Pasek and colleagues in Psychological Science examined the design features of top-selling educational apps for children and found that most apps failed to incorporate any of the four pillars of learning science that predict durable learning: active learning, engagement with the material, meaningful context, and social interaction. They were engaging — children liked them — but engagement and learning are not the same thing.

The study did not condemn apps categorically. It identified the subset of apps with learning-science-aligned design features and found those apps did produce measurable learning. But they represent a small fraction of the market. For parents buying apps based on the “educational” label in the App Store, the research offers little reassurance.

Technology Types and Evidence Strength

Technology Type	Claimed Benefit	Evidence Quality	Evidence Verdict	Best-Case Conditions
Tablets (1-to-1 programs)	Improved engagement, personalized learning	Moderate (multiple RCTs)	No consistent academic gain; engagement benefit only	Supplemental, structured use with teacher oversight
Interactive whiteboards / smartboards	Better instruction, higher engagement	Moderate (UK/US studies)	No academic advantage over standard whiteboards	Quality of lesson matters, not the board
Educational apps (consumer)	Skill building, early literacy/math	Low (largely observational)	Most fail learning-science design standards	Rare apps with active, contextualized design
Adaptive learning software (math)	Personalized practice, fluency	Moderate-High (some RCTs)	Modest positive effects in math fluency	Supplemental, 2–3x/week at recommended dose
Online video instruction (Khan Academy)	Concept explanation, pacing	Low-Moderate	Mixed; no consistent outcome advantage over live instruction	Self-directed remediation, student-initiated use
Learning management systems (Schoology, Canvas)	Organization, communication	Low (no outcome RCTs)	No academic outcome data; organizational benefit only	Consistent teacher implementation
Typing / keyboarding software	Writing fluency, output speed	Moderate	Modest positive effects on writing output for middle school	Integrated with writing instruction
Coding platforms (Scratch, Code.org)	Computational thinking, engagement	Low-Moderate	Positive engagement and CS interest; limited academic transfer	Project-based, teacher-facilitated

What to Actually Do

The research doesn’t call for abandoning technology in classrooms. It calls for using it as one tool among many, with realistic expectations about what it can and can’t do, and with attention to the conditions under which it actually helps.

Ask the Right Questions About Your School’s Tech Adoption

When your child’s school announces a new technology initiative — a one-to-one iPad program, a new learning platform, a smartboard rollout — you’re entitled to ask what evidence the district used to make the decision. Specific questions worth raising at PTO or school board meetings:

• Was the evaluation conducted by an independent researcher or by the vendor?
• What was the comparison condition — did students using this technology do better than similar students who didn’t?
• What training are teachers receiving, and how is implementation being monitored?
• How will the district know in two years whether this worked?

Districts that have done their homework will have answers. Districts that bought based on vendor presentations often won’t. Both outcomes are useful information.

Treat Technology as a Supplement, Not a Replacement

The research is consistent on this point: technology that supplements good instruction produces some benefit in specific areas. Technology that replaces direct instruction generally does not. If your child’s teacher is using an app or platform primarily to occupy students while the teacher addresses other groups, that’s an engagement strategy — not an academic intervention. The research doesn’t support it as a learning tool in that deployment.

This doesn’t mean teachers who use tech for management are doing something wrong. It means parents should be calibrated about what that use does and doesn’t accomplish academically.

Evaluate the Apps Your Child Uses at Home

If you’re paying for educational apps or allowing your child to use them as learning time, the Hirsh-Pasek et al. (2019) framework gives you four design features to look for:

1. Active learning: Does the app require the child to do something meaningful, or just tap and watch?
2. Engagement with the content: Is the interaction related to the learning goal, or is the animation/game rewarding for reasons unrelated to the skill?
3. Meaningful context: Is the skill embedded in a meaningful task or just isolated drill?
4. Social interaction: Does the app encourage conversation with a parent or peer about the content?

Apps that satisfy all four features are rare but exist. Apps that fail all four are common and dominating the best-seller lists. Spending twenty minutes on a worksheet with a parent present likely outperforms thirty minutes on a failing-criteria app — and costs nothing.

Don’t Interpret Technology Enthusiasm as Evidence

Children almost universally prefer using tablets over pencils and workbooks. This preference is real, strong, and entirely uncorrelated with learning outcomes in the research. A child who is engaged, motivated, and enjoying an app may be learning nothing beyond how to play the app more efficiently. Enjoyment is a legitimate goal. It’s just not the same goal as learning, and conflating them has driven billions of dollars of ineffective technology spending.

If your child finishes a forty-minute session on an educational app excited and energized, ask her a question about the content — not about the game. The answer often tells you more than the engagement metric does.

Focus on Teacher Quality, Not Tool Quality

The research on what predicts classroom learning outcomes consistently finds teacher quality as the dominant variable — larger than class size, curriculum, or technology. The Moss et al. smartboard studies found that teacher skill explained outcome variation far better than whether a smartboard was present. A skilled teacher with chalk produces better outcomes than a mediocre teacher with a smartboard. This isn’t a counsel to oppose technology. It’s a counsel to invest parental advocacy energy in the quality of instruction, not the sophistication of the tools.

What to Watch for Over the Next 3 Months

If your child’s school recently adopted or expanded a technology platform, you have a natural observation window over the next semester.

Week 4: Ask your child to explain something she learned on the school’s tablet or app platform — not the mechanics of the app, but a concept or skill. If she can explain it clearly, the tool is working at some level. If she can describe the game but not the learning, the engagement is real but the learning transfer may not be.

Month 2: Look at homework and classwork for evidence of the skills the platform is meant to develop. If the school is using an adaptive math app, are you seeing improvement in math homework completion and accuracy? If not, ask the teacher whether she’s seeing progress data from the platform — most adaptive tools generate detailed usage reports — and what it shows.

Month 3: At the semester mark, look at any formal assessment data — unit tests, reading level assessments, interim assessments — and ask whether the trend matches what the technology provider predicted. Districts that bought a platform with promises of X months of learning gain per year should be tracking against that claim. If nobody is tracking it, that’s worth raising.

Frequently Asked Questions

Are tablets in classrooms harmful to kids?

Not inherently. The research doesn’t show tablets cause harm in most conditions. What it shows is that tablets rarely produce the academic gains they’re marketed to produce, and that excessive screen use during instruction time can crowd out reading, writing, and direct instruction time that does have a stronger evidence base. The harm risk is primarily one of opportunity cost — time spent on low-evidence tech use is time not spent on higher-evidence instruction.

Do educational apps teach kids to read?

Some reading apps with evidence-based phonics instruction — particularly those aligned to the science of reading — show genuine benefits for early literacy. These typically include systematic phonics sequences, decodable text practice, and explicit phonemic awareness work. Generic “reading” apps that focus on story engagement or letter recognition without systematic phonics instruction have minimal evidence of reading improvement.

Why do teachers love smartboards if research says they don’t help?

Teacher enthusiasm for smartboards is real and documented. The tools are genuinely easier to use for certain presentations, allow teachers to save and share lesson materials, and enable multimedia integration. These are real pedagogical conveniences. They just don’t appear to translate into student outcome improvements in controlled studies. Teachers can love a tool and be right that it’s convenient while researchers are also right that it doesn’t move the outcome needle. Both things are true.

My child’s school bought 200 iPads. Should I be concerned?

Only if the purchase replaced investment in teacher development, curriculum quality, or instructional time. A district that buys iPads on top of strong existing instruction may get modest supplemental benefits. A district that buys iPads as a substitute for addressing teacher quality or curriculum gaps is likely to be disappointed by the outcome data in two to three years.

What classroom technology actually has evidence behind it?

Adaptive math software used at recommended dosages, systematic phonics software with decodable text for early readers, and typing instruction software for middle school writing output all have modest but real evidence of benefit. These are narrow applications — not the “transformational learning” the marketing promises, but real tools for specific jobs.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

Sources

1. OECD. (2015). Students, Computers and Learning: Making the Connection. PISA, OECD Publishing. https://doi.org/10.1787/9789264239555-en
2. Cheung, A. C. K., & Slavin, R. E. (2013). The effectiveness of educational technology applications for enhancing mathematics achievement in K-12 classrooms. Educational Research Review, 9, 88–113.
3. Moss, G., Jewitt, C., Levaãiç, R., Armstrong, V., Cardini, A., & Castle, F. (2007). The Interactive Whiteboards, Pedagogy and Pupil Performance Evaluation. Institute of Education, University of London.
4. Zheng, B., Warschauer, M., Lin, C. H., & Chang, C. (2016). Learning in one-to-one laptop environments: A meta-analysis and research synthesis. Review of Educational Research, 86(4), 1052–1084.
5. Hirsh-Pasek, K., Zosh, J. M., Golinkoff, R. M., Gray, J. H., Robb, M. B., & Kaufman, J. (2015). Putting education in “educational” apps: Lessons from the science of learning. Psychological Science in the Public Interest, 16(1), 3–34.
6. Pane, J. F., Steiner, E. D., Baird, M. D., & Hamilton, L. S. (2015). Continued Progress: Promising Evidence on Personalized Learning. RAND Corporation.
7. What Works Clearinghouse. (2016). DreamBox Learning Math (Grades K–8). U.S. Department of Education, Institute of Education Sciences.
8. Dolan, J. E. (2016). Splicing the divide: A review of research on the evolving digital divide among K–12 students. Journal of Research on Technology in Education, 48(1), 16–37.
9. Roschelle, J., Feng, M., Murphy, R. F., & Mason, C. A. (2016). Online mathematics homework increases student achievement. AERA Open, 2(4), 1–12.